18.08.2020
Positive Psychology and Technology (M.Sc.) 20 EC Master Thesis
Interventions for Cognitive Impairments and Mental Health by the Means of Virtual Reality among Stroke Survivors
-
A Systematic Review
Carina Rebecca Behle
First Supervisor:
Dr. Christina Bode
Second Supervisor:
Dr. Pelin Gül
1
Abstract
Background: Cognitive impairments and poor mental health are prominent adverse effects that patients are often burdened with after suffering from a stroke. Treatment guidelines are not available yet for the two different factors. For that reason, Virtual Reality is explored as a possible intervention as it can take on different modes, be applied in different settings, and hence is quite adjustable to a stroke patient’s special needs. The objectives of this review were to examine characteristics of Virtual reality interventions aimed at improving a) cognition and b) mental health in stroke survivors at home and inpatient/outpatient settings and their
effectiveness. Further, similarities and differences between the different settings were explored.
Method: The data sources used for this systematic literature review were Scopus, Web of Science, and PubMed. The main eligibility criteria concerned records with stroke-patients, virtual reality interventions, outcomes concerning cognition and/or mental well-being as well as a publication after 2016. Both, qualitative and quantitative data were reviewed.
Results: A total of 1824 records were found, of which 23 were used. During the review process next to the home and inpatient setting, a mixed setting was identified and added. Only a small number of studies made use of fully immersive virtual reality in all settings. Instead of commonly known VR glasses, tablets and computer working stations were mainly used. The most prevalent intervention mode consisted of games. Results on the effectiveness of VR deviated among the records, still it is seen to be beneficial as a complementary treatment component to conventional treatment.
Conclusion: Limitations of this review may lie in its single-rater nature and the variety of research designs included. Implications for future research are to further explore the use of VR for mental health interventions in different settings as well as to explore a possible
integration of different settings, to smoothen the transition after discharge, and simultaneously
allow a continuation of rehabilitation. Overall, VR seems to be a promising complementary
intervention for cognitive impairments and poor mental health among stroke patients.
2
Content
Abstract ... 1
Introduction ... 3
Cognitive Consequences of Stroke ... 3
Mental Health Consequences of Stroke ... 4
Interventions ... 4
Virtual Reality... 4
Clinical vs. Home Setting ... 7
Current Study ... 8
Methods ... 9
Search Strategy ... 9
Study selection ... 9
Data analysis ... 10
Quality Assessment ... 10
Results ... 11
Selection of studies ... 11
Identification ... 12
Screening ... 12
Eligibility ... 12
Included ... 12
Quality Assessment of studies ... 13
Population and Study Design Characteristics ... 14
Intervention characteristics ... 17
Intervention design ... 19
Study outcomes ... 21
Cognition ... 21
Mental health ... 22
Similarities and differences ... 23
Differences ... 23
Similarities ... 25
Discussion ... 26
Strengths and limitations of this study ... 29
References ... 31
Appendices ... 37
Appendix A ... 37
Appendix B ... 38
Appendix C ... 39
Appendix D ... 41
3
Introduction
Stroke is seen to be the second leading cause of death and the third major cause of disability across the globe (Feigin et al., 2015). The development of treatments during the acute phase has led to a decrease in mortality among stroke patients. However, many are left behind with physical, cognitive, and psychological impairments. Between the years 1990 and 2010 an increase of 86.7% could be noticed in global years lived with disability (YLDs) among stroke survivors (Vos et al., 2012). Even though many rehabilitation programs have been designed and shown to be successful in the treatment of physical impairments, the same does not hold true for the restoration of cognitive functions. As indicated by Kapoor (2017), about half of stroke survivors, despite rehabilitation, had cognitive impairments two to three years after the onset of the stroke. In terms of mental health, 20-30 percent of stroke survivors suffer from symptoms as for example depression or anxiety post-stroke (Kirkevold et al., 2018).
Cognitive Consequences of Stroke
When a stroke occurs, the arteries leading towards and within the brain, are affected (American Stroke Association, 2020). The blood vessels that supply the brain with oxygen and nutrients may be blocked or ruptured. In such cases, the brain cells of the certain regions die due to the undersupply caused by the affected vessel.
Since a stroke is causing damage to certain areas of the brain it is standing to reason that a patient’s cognitive functions are affected as well. Impairments in cognitive functioning might affect only one cognitive area, but as indicated by Kim et al. (2011), they are more likely to affect several areas of cognition at the same time such as ”attention, concentration, memory, spatial processing skills, language, problem-solving skills, and planning skills”.
Research has shown that approximately two-thirds of stroke survivors suffer from cognitive
impairments 2 and 3 months after the onset of stroke (Nijsse et al., 2017; Blackburn et al.,
2012). About 50% of people are still experiencing cognitive restrains 6 months post-stroke
and despite rehabilitation even after 2 to 3 years (Nijsse et al., 2017; Kapoor, 2017). As
indicated by Cumming et al. (2014), cognitive impairments might be associated with a poorer
quality of life post-stroke. Further, poor cognitive functioning, regardless of the physical
functioning, increases the likelihood of being dependent on other people’s aid after discharge,
may it be at home or in nursing facilities (Tatemichi et al., 1994).
4 Mental Health Consequences of Stroke
Apart from the rather obvious impairments due to stroke, such as physical and
cognitive dysfunctions, stroke may also be related to a client’s mental health. As indicated by Norrving et al. (2018), “around a third of stroke survivors are disabled, have poor post-stroke cognitive ability and poor mental health”. Different factors can be representative of poor mental health among stroke survivors, such as depression, anxiety, general psychological distress, and fatigue, just to name a few (Kirkevold et al., 2018; Norrving et al., 2018).
Experiencing such difficulties may influence a patient's long-term functioning, impede the rehabilitation process, and impact one's overall quality of life (Kirkevold et al., 2018). To give some examples with regards to the prevalence of mental health issues post-stroke, as indicated by the authors Kirkevold et al. (2018), about one-third of the patients experience depressive symptoms post-stroke, which might lead to a full depression that may still be present several years post-stroke. When it comes to anxiety, approximately 20% of stroke survivors
experienced anxiety within the first months post-stroke.
Interventions
Until today, treatment guidelines of stroke solely focus on the recovery process of physical impairments. To give examples, the European Stroke Association (2008) and the American Stroke and Heart Organization (Powers et al., 2018) mentioned that cognitive functions and mental health can be affected by a stroke but that research data is not sufficient to provide enough evidence for effective treatment. However, due to the high numbers in the prevalence of cognitive impairments and poor mental health one might wonder what kind of interventions, respectively treatments, are available and applied at the moment. For cognitive impairments in general, one can say, that two different forms of cognitive rehabilitation exist in today's clinical application (de Luca et al., 2018). The first being restorative rehabilitation, which facilitates the development of lost functions by the means of cognitive exercises. The latter type is named compensatory rehabilitation, which trains the patient in making use of aids and tools. The progress in the advancement of technologies in the health sector has also made its way into rehabilitation bringing along computer-based interventions. One of those interventions entails the application of Virtual Reality (VR).
Virtual Reality
With the aid of VR, patients can run through both of the aforementioned forms of
cognitive rehabilitation in “ complex individualized and natural simulated environments”
5
(Kober et al., 2013). Not only can VR be used for cognitive rehabilitation but the assessment and treatment of mental health as well (Valmaggia et al., 2016). VR can be applied in
different forms and since there is an ongoing development in the field of technology, VR does also keep on developing at a rapid pace. VR can be described as a virtual environment that a person can find him- or herself in and can interact with. A difference can be made between immersive, non-immersive and semi-immersive VR. Among immersive technology,
“headmounted displays, body movement sensors, real-time graphics, and advanced interface devices (e.g., specialized helmets)” are used (Zeng et al., 2018). The user can turn around
360° within the virtual environment, which can generate a stronger sense of presence within that environment (Ventura et al., 2016) and hence might increase the sensation of reality. Utilizing non-immersive technology, the user rather functions as an observer from the outside, still able to enact with the environment by moving connected devices. The interface of non-immersive VR rather consists of devices such as flat screens “and requires the use of a corresponding keyboard, controller, and/or joystick” (Quian et al., 2020). Overall, one can say, that the degree to which a user feels actually present in the virtual environment decreases with the degree of immersiveness (Zeng et al., 2018).
To name a few more different hardware devices, which are capable nowadays to display virtual environments, those can be tablets, smartphones, computers, or VR glasses.
Numerous software exist that run on those hardware devices. Further, those VR modes can take on different forms such as exercises, serious games, or exergames. Serious games comprise of those games that do not solely serve the purpose of entertainment but aim at e.g.
educating the player or benefitting one's health. Another game-mode would be exergames which emphasize body movement through for example movement-tracking. To create an impression of what tasks in the virtual environment might look like, they can be designed for the patient to implement things such as grocery shopping at the supermarket or other daily activities.
Additionally, VR is also being tested for its applicability not only in treatment but as an assessment-tool for example for mental health (Bell et al., 2020). Bell et al. (2020) concluded that VR can be beneficial as an assessment tool in a broad range among mental health. The authors stated that “VR elicits similar psychological and physiological reactions to real-world environments, extending the reach of current assessments beyond the lab or clinic”
(Bell et al., 2020). According to them, VR could be beneficial in mental health assessment as
it comes closer to real-life experiences, which cannot be fully given in laboratories or clinics.
6
As described by Kim et al. (2011), stroke survivors can work in VR independently of any physical disabilities such as motor-dysfunctions of their limbs or hands, which adds a great advantage to it. This way, cognitive abilities can still be trained, and mental health can be improved, even though patients might not be able to visit outpatient centers or go grocery shopping on their own. VR can establish a safe and guided virtual environment, which might be relocated into the real-life again, as implicated by Chen et al. (2020). Those applications offer a chance for stroke survivors, to rehearse activities not fearing to do something wrong (Chen et al, 2020).
However, one should take a stroke survivors special needs into account as well. Visual impairments as for example a visual field loss, that is common among stroke survivors and other deficits in perception and senses might occur (Wiley et al. 2019), asking for very user- specific characteristics. Since VR mainly runs employing visual stimuli, this tool option might not be fully suitable for stroke patients with visual limitations or it would need to be
sufficiently adaptable.
As indicated by Wiley et al. (2019) and Kannan et al. (2019), semi-immersive and non-immersive VR might be more appropriate, due to economical factors. High-quality VR glasses might be found at the upper end of price ranges until this point, whereas tablets, computers, and smartphones are more likely to already be owned by patients. Beyond that, semi-immersive and non-immersive VR devices have a smaller degree of intrusiveness, which might be rather suitable for stroke survivors ( Wiley et al. 2019). The authors mention that the glasses that are used are heavy. The heavyweight might be unfavorable for patients as their physical and strength capacities might have been enfeebled by the stroke. Additionally, among longer use, unfavorable side-effects such as eye strains may occur (Wiley et al., 2019).
Even among healthy patients, cybersickness can occur, which derives from a visual-vestibular conflict (Kim et al., 2018). This conflict can emerge once the actual movements of a user do not match the movement in the virtual environment. As a consequence, adverse effects such as nausea, dizziness or eye strains can burden the user. Since stroke survivors often suffer from cognitive deficits and especially visual field deficits (Pollock et al., 2011), they might be even more affected by the usage of VR devices.
Several reviews have shown, that VR “can be a promising and effective tool in the
recovery of neurological symptoms, including cognitive ones” (Maggio et al., 2019). A
review by Moreno et al. (2019) indicated that VR in cognitive rehabilitation among stroke
patients might improve memory, dual tasking, and visual attention. Further, VR treatment has
7
shown to be effective among stroke patients especially in the area of visual-spatial functioning, (Dehn et al., 2020) and attention (De Luca et al., 2018).
When it comes to a patient’s mental health, conditions such as post-stroke depression are often neglected in treatment, even though they might have an impact on the physical recovery and a patient's quality of life. As indicated by Paolucci (2008) only a minority of patients receives a diagnosis and gets treated with regards to their depression. According to Kirkevold et al. (2018), many studies have been conducted, aiming to explore potential interventions that prevent or treat post-stroke psychological problems, but the outcomes have not been promising yet. A qualitative study conducted by Chen et al. (2020), however,
indicated that VR under the form of telerehabilitation might be beneficial for the improvement of mental wellbeing in stroke patients. Chen et al. (2020) reported that the patients perceived subjective positive effects with regards to their limb functions, cognition, and emotional well- being due to the telerehabilitation. Further, it was mentioned, that an educational module of the intervention helped participants to understand and cope with their misfortune.
Clinical vs. Home Setting
Even though treatment options seem to be available and VR a promising method in the field of rehabilitation after stroke, the numbers of stroke survivors, suffering from cognitive deficits and mental problems still seem to be quite high, especially when taking a look at the prevalence after a couple of years. As indicated by Sheehy et al. (2019), most stroke patients are discharged from inpatient rehabilitation after 8-10 weeks after the stroke. Afterward, for most rehabilitation is still necessary, but not necessarily easy to take part in. This could be explained by the fact that outpatients still might suffer from continuing physical deficits accompanied by dependence and transportation difficulties.
In other areas, especially in the rehabilitation of physical impairments such as upper limbs, a multitude of quantitative research about the application and effectiveness of VR in the home environment has already been conducted and systematically reviewed,
demonstrating the possibility of VR as a treatment option in home settings in general (Schröder et al. 2018). Results revealed, that an increase in motivation might be implied by VR training sessions due to the game mode. Similar results were found by Llorens at al.
(2016), who reported that a competitive game mode combined with virtual reality can be beneficial for motivation among stroke patients with cognitive deficits. As a consequence, a higher motivation might result in patients engaging in longer training sessions more
frequently, which could benefit the overall rehabilitation outcome.
8
Functional improvements appear to be comparable to the progress in clinical settings, also emphasizing the factor of cost-efficiency. However, for cognitive impairments and mental health, no systematic reviews exist in the home setting. There might lie several advantages in introducing at-home VR-treatment for stroke survivors. Those might be decreasing the necessity of transportation, as the mobility of stroke survivors still might be affected. Among the issue of mobility comes also the factor of dependence, which might be decreased as well, once treatment can be conducted from home. Further, economical aspects come into play as well, that might be counteracted by a simplification of the effort put into treatment by several parties.
Current Study
To summarize, many patients’ cognitive functions and mental health are affected by a stroke. Not only do those factors seem to impede the rehabilitation process of physical impairments, but they can affect stroke survivors’ quality of life. VR is one way of treating cognitive impairments and mental health problems. However, this kind of intervention is not necessarily kept on after being discharged from clinics or rehabilitation centers. Since cognitive impairments and mental health problems have shown to be still present a couple of years post-stroke it is of great interest to investigate how interventions of applying VR at home might contribute to the improvement of those factors among stroke survivors and hence their overall quality of life. Until today, most research has focused on the application of VR in clinical rather than home settings when it comes to treating cognitive impairments and mental problems among stroke patients. As far as our knowledge goes, no systematic reviews exist yet, exploring the characteristics as well as the effectiveness of VR interventions for stroke survivors in their home environment. The recent systematic reviews focus on physical
rehabilitation, but not on cognitive and mental interventions. For that reason, the objectives of this research are as follows:
1) to examine characteristics of Virtual Reality interventions aimed at improving a) cognition and b) mental health in stroke survivors at home and inpatient/outpatient settings
2) to examine the effectiveness of Virtual Reality interventions aimed at improving a) cognition and b) mental health among stroke survivors at home and inpatient/ outpatient settings
3) to examine similarities and differences with regards to characteristics and effectiveness of
Virtual Reality interventions between the in-home and inpatient/outpatient settings.
9
Methods
Search Strategy
The current study followed the guidelines of a systematic literature review in the form of a state-of-the-art review. This type of methodology appears suitable as it covers more temporary matters (Higgins & Green, 2008, as cited in Grant & Booth, 2009). Especially in the field of VR, focussing on more recent findings of the past few years appears adequate, because of the fast development of the technology and its rising prevalence in usage in and outside of research.
Due to the difference in the nature of the objectives of this study, a qualitative, as well as quantitative synthesis, will be conducted. To create a general overview and identify
characteristics of possible and/ or available interventions, a qualitative synthesis will be conducted. Possible quantitative results of the former qualitative synthesis will be included to evaluate the effectiveness of the identified interventions.
In this thesis, the literature search was conducted in the electronic databases Scopus, PubMed, and Web of Science. The databases Scopus and Web of Science were chosen due to their multidisciplinary content that could combine the technological, medical as well as psychological aspects of this study. The database PubMed was added due to the medical component of this study. Within the databases, it was searched for published journal articles.
The following search terms were used: stroke AND “virtual reality” AND cognition AND/OR “mental health”. The full search strategy can be found in Appendix A. The search was conducted in April 2020.
Study selection
In line with the guidelines of Higgins & Green (2008 as cited in Grant & Booth, 2009), the articles were selected employing the following steps. First, articles were detected by searching in the database. In the second step, titles and abstracts were examined. In the third step, the full texts of the relevant reports were retrieved. In the next step the full texts were screened and in the final step accepted or rejected based on the inclusion and exclusion criteria.
To be included in this review, the following inclusion criteria had to be met for the
qualitative synthesis: 1) the sample included stroke patients, 2) the study included VR as an
intervention, 3) publication in English or German language 4) publication in a peer-reviewed
10
journal, 5) published after 2016 for the home setting 6) published after 2018 for the
inpatient/outpatient setting. The publication year after 2016 for the home setting was chosen due to the state-of-the-art nature of this review, as it supposes that one should only focus on the last couple of years. Since systematic reviews have been published until 2018 in the inpatient/outpatient setting, it was decided to set the year 2018 as an inclusion criterion to prevent any repetition of former work and instead create novel insights. For reports to be eligible for the quantitative synthesis 7) the study design has to be quantitative. Additionally, 8) the research design of the inpatient-reports had to be either in the form of a literature review or randomized controlled studies due to the multitude of records.
Excluded from the review were reports: 1) not including stroke as a medical condition, 2) where outcomes concerning stroke were not explicitly reported, if other medical conditions were included, 3) published before 2017 for the home setting, 4) published before 2019 for the inpatient/outpatients setting. An explanation for the exclusion criteria of the publication year choice is identical with the inclusion criteria decision.
Data analysis Quality Assessment
The Mixed Methods Appraisal Tool (MMAT)- Version 2011 (Pluye et al., 2011) was used for the quality assessment. The rationale behind this tool choice can be explained by the broad scope of study designs included in this review. The MMAT allows to include
qualitative, quantitative (randomized controlled, non-randomized, descriptive) and mixed- method designs. The tool comprises of different items, which need to be answered. The first two items apply to every study design. In case the first two items are not successfully met, further appraisal should not follow due to insufficient scientificity. The aforementioned two items are followed by four items specific to the studies design, except for the mixed methods design, which uses three follow-up items (See Appendix B for the content of the items). With regard to the scoring metrics, an overall quality score can be calculated. For each criterium, the score is indicated by scores from 1-4. The total score for qualitative and quantitative studies is calculated with the sum of scores divided by four, indicating the number of criteria met (25%-100%) and therefore the quality of the study. Concerning the mixed methods studies, “the overall quality of a combination cannot exceed the quality of its weakest component” (Pluye et al., 2011).
Due to the possible diversity of study designs, an additional appraisal tool will be
included for systematic reviews, namely the methodology checklist for systematic reviews
11
and meta-analysis provided by the Scottish Intercollegiate Guidelines Network, Healthcare Improvement Scotland (2019, see Appendix C for the entire checklist) . Systematic reviews with or without meta-analysis will be included in this review.
Data Extraction
The following data were extracted from the included articles:
• characteristic of the study design o study design
o randomization/blinding/control groups
• Characteristics of the population o country
o diagnosis o sample-size o time post-stroke
• characteristics of the intervention o setting (home/inpatient/other) o treatment objective
o intervention mode (e.g. game. Exergame, exercise, education) o hardware/software
o delivery mode (number of sessions, duration, assistance available) o outcome measurements of interest for this review (primary, secondary
outcome, such as cognitive functioning, mental well being in general or indicated by e.g. depression, fatigue, anxiety)
• Outcomes
o Measurement tools o Results
o Conclusion
Results
Selection of studies
By using the databases PubMed, Scopus, and Web of Science a total of 2807 records
were retrieved in the literature search (see Figure 1). The number of duplicates, which were
removed, amounted to 983, and 1197 records published before the year 2017 were excluded
as well due to the nature of this state-of-the-art study. The remaining 627 records were
12
screened on title and abstract, resulting in the exclusion of further 464 records. A total of 163 full-text articles were screened of which 23 articles remained as eligible for the qualitative synthesis (Objective 1) and 17 for the quantitative synthesis (Objective 2). During the literature review, a third group was identified that appeared to be essential, consisting of studies that combined inpatient/outpatient or in-home rehabilitation.
Figure 1- Flowchart of the Literature Search
In clud e d Eli gibili ty
Scopus Cognition (n=1086) mental health (n =359) Total- duplicates (n=1286)
Id e n tific ati o n Scre e n ing
Records after duplicates removed (n =1824)
Records screened (n = 627)
Records excluded (n =464)
Full-text articles assessed for eligibility
(n = 163)
Full-text articles excluded, with reasons
(n = 140 )
Studies included in qualitative synthesis
(n = 23 )
Studies included in quantitative synthesis ()
(n = 17 )
Web of Science Cognition (n=868) mental health (n =224) Total- duplicates (n=987) PubMed
Cognition (n=534) mental health (n =40) Total- duplicates (n=534)
Records excluded since published before 2017
(n = 1197 )
13
In the following, the quality of studies, their characteristics with regard to the study population, design, intervention, and outcomes are presented. Findings will be presented organized by the research objective and further divided by the setting it was designed for.
Quality Assessment of studies
The quality of the 21 qualitative and quantitative records were appraised by means of the Mixed Methods Appraisal Tool (MMAT) and the quality of the two literature reviews by Wiley et al. (2019) and Maggio et al. (2019) by means of the Checklist for Systematic Reviews and Meta-analysis provided by the Scottish Intercollegiate Guidelines Network, Healthcare Improvement Scotland (2019) (See Table 1 and 2). The majority of the quality of records was highly satisfying, ranging from 2-4, indicating 75%-100% satisfaction. Four out of 21 studies scored at 2 or less, indicating a less satisfying quality of fulfilling only 50% or less of applicable characteristics. Three Records did not receive a total score due to their nature of being protocols. However, based on the given evaluation points, the quality would appear satisfying as well. With regards to the literature reviews, the review by Wiley et al.
(2019) was evaluated to have high quality, whereas the review by Maggio et al. (2019) was rated with low quality (see table 2).
Table 1- Quality Assessment- Qualitative, Quantitative and Mixed-methods
Author. 1 2 .1 .2 .3 .4 Total
Burdea et al., 2019 √ √ 2 4 2 - 2
Calabro et al., 2018 √ √ 4 3 - - -
Cano-Manas et al., 2020 √ √ 4 3 3 3 3,25
Chen et al., 2020 √ √ 3 4 4 4 3,75
Dodakian et al. 2017 √ √ 2 4 3 3 3
Fu et al., 2019 √ √ 4 3 4 2 3,25
Kannan et al., 2019 √ √ 3 1 2 1 1,75
Kilbride et al., 2018 √ √ 3 4 2 - -
Lee et al., 2020 √ √ 4 3 4 4 3,75
Lin et al., 2020 √ √ 4 3 4 4 3,75
Maier et al., 2020 √ √ 4 2 4 3 3,5
Maresca et al., 2019 √ √ 1 1 1 1 1
McKay et al., 2019 √ √ 4 4 - - -
Mohd et al., 2019 √ √ 2 2 3 1 2
14
Table 1 continuedAuthor. 1 2 .1 .2 .3 .4 Total
Oh et al., 2019 √ √ 4 3 4 4 3,75
Paulino et al., 2019 √ √ 1 3 4 1 2,25
Pugliese et al., 2019 √ √ 3 4 3 4 3,5
Rai et al., 2017 √ √ 3 3 2 2 2,5
Rivas et al., 2019 √ √ 3 3 3 1 2,5
Torrisi et al., 2019 √ √ 4 3 1 1 2,25
Wall et al., 2018 √ √ 3 3 4 2 3
Column 1 and 2 represent the screening questions for all type with a checkmark √ indicating the item to be answered with yes; Columns with .1,.2,.3,.4, refer to the 4 follow-up items of the different study designs; The last column represents the total score of the paper
Table 2-Quality Assessment literature review
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 2.1 2.2
Wiley √ √ √ √ √ x √ √ √ √ x High
quality
√
Maggio √ √ Can´t say
Can´t say
x x √ x x x √ Low
quality
√
Checkmark √ indicates the item to be met; Cross x indicates the item not to be met
Population and Study Design Characteristics
The entire data of the population characteristics are displayed in Table 3. The studies included in this review emerged all over the world including Europe (Italy, Spain, Portugal, UK), America (USA, Canada, Mexico), Asia (Malaysia, Taiwan, South Korea), and Australia.
The number of (expected) participants ranged from 5-152. Nineteen out of the 23 studies focused solely on stroke as a diagnosis, two studies added Traumatic Brain injury and infection and the remaining two focused on either Aphasia post-stroke or Hemiplegia post- stroke. The age of participants reached from 37 to over 90 years with one exception of a study implemented with children aged 4-7. When it comes to the recruitment methods, different approaches could be seen such as convenience sampling, community posters, newsletters, recruitment through therapists/doctors. The time passed after the stroke until the
implementation of the intervention would range between 3 days and 10 years.
15
Table 3- Population Characteristics
Author Country Sample Size Diagnosis Time post-stroke Setting Burdea et al.,
2019
USA N=7; Stroke Stroke >9 months Home
Calabro et al., 2018
Italy N=80 Stroke (n=40) <1 year home
Chen et al., 2020
USA N=13 Stroke 4-36 weeks Home
Dodakian et al., 2017
USA N=12 Stroke 3-24 months Home
Kilbride et al., 2018
England N=30 Stroke >12 weeks Home
Maier et al., 2020
Spain N=30 Stroke 6months<x<10years Home
McKay et al., 2019
Australia N=80 ABI (TBI,
Infection, Stroke)
>6 months Home
Mohd et al., 2019
Malaysia N=6 Stroke - home
Paulino et al., 2019
Portugal - home
Rai et al., 2017 Australia N=5 Stroke - home
Rivas et al., 2019
Mexico N=5 Stroke - home
Cano-Manas et al., 2020
Spain N=56 Stroke 15days-6monts Inpatient
Kannan et al., 2019
USA N=24 Stroke > 6 months Inpatient
Lin et al., 2020 Taiwan N=152 Stroke 24 hours-3 days Inpatient Maggio et al.,
2019
Italy N=112 for Stroke;- N=25 for Control
Stroke - inpatient
Oh et al., 2019 South Korea
N=31 Stroke 6 months Inpatient
Wiley et al., 2019
Canada N=124 Stroke 18.2 ± 11.3 days to
41.1 ± 41.0 months
Inpatient
Fu et al., 2019 USA N=7 Post-stroke hemiplegia
>6 months Outpatient &
Home Lee et al., 2020 South
Korea
N=36 Stroke < 6 months Inpatient,
outpatient Maresca et al.,
2019
Italy N=30 Post-stroke
Aphasia
- Inpatient &
home
16
Table 3 continuedAuthor Country Sample Size Diagnosis Time post-stroke Setting Pugliese et al.,
2019
Canada N=30 Stroke 4 days Inpatient &
home Torrisi et al.,
2019
Italy N=40 Stroke 3-6 months Inpatient &
home Wall et al.,
2018
Australia N=96 Stroke - Community-
dwelling
With regards to the study design (see Table 4), a total of five Randomized Controlled Trials (RCT), two RCT pilot studies, two RCT study protocols, and two systematic reviews were included. Further five study designs took the form of Intervention Technology design studies, a feasibility study, a feasibility study protocol, a qualitative semi-structured interview, a system performance pilot study, a single group prospective cohort study, and a controlled parallel-group pilot study.
Table 4- Study Design Characteristics
Author Study Design Setting
Burdea et al., 2019 Feasibility Study Home
Calabro et al., 2018 Study protocol of a multicentric observational, rater-blinded, active-controlled,
parallel-group pilot study
home
Chen et al., 2020 Qualitative, semi-structured interviews Home Dodakian et al.,
2017
System performance pilot study Home
Kilbride et al., 2018 Study protocol about Intervention feasibility nonrandomized trial
Semi-structured interviews
Home
Maier et al., 2020 RCT pilot study Home
McKay et al., 2019 Study protocol of RCT Home
Mohd et al., 2019 Intervention technology Design & Feasibility study concept home Paulino et al., 2019 Intervention technology redesign study home Rai et al., 2017 Intervention Technology design study home Rivas et al., 2019 Intervention technology development study home Cano-Manas et al.,
2020
RCT Inpatient
Kannan et al., 2019 RCT Inpatient
17
Table 4 continuedAuthor Study Design Setting
Lin et al., 2020 RCT Inpatient
Maggio et al., 2019 Systematic review inpatient
Oh et al., 2019 RCT Inpatient
Wiley et al., 2019 Systematic review and Meta-Analysis Inpatient
Fu et al., 2019 Exploratory case study series Outpatient &
Home
Lee et al., 2020 RCT Inpatient,
outpatient
Maresca et al., 2019 RCT pilot study Inpatient & home
Pugliese et al., 2019 unblinded, single-group, prospective cohort study Inpatient & home Torrisi et al., 2019 Prospective, assessor-blinded, controlled parallel-group pilot
study
Inpatient & home
Wall et al., 2018 Intervention Technology design study Community- dwelling
I ntervention characteristics
In the following, the characteristics of the VR interventions aimed at improving a) cognition and b) mental health in the different settings will be presented in line with the first research objective. The characteristics are based on different criteria as displayed in Tables 5 and 6 (for detailed information, see Appendix D).
Technology
Home. From 10 out of the 11 studies focusing on the home setting, only one study applied a fully immersive VR, namely the study conducted by Mohd et al. (2019). In their study, the researchers made use of the HTC Vive virtual reality headset. However, the authors indicated as well, that the intervention is designed to be implementable with other, more affordable devices to maximize consumer-friendliness. The remaining 10 records reported only methods of a non-immersive nature.
The majority of technical devices creating virtual environments consisted of
workstations build of Computers and Screens. Apart from that, two studies implied the use of
tablets and the study by Rai et al. (2017) the use of any mobile device with a web browser. As
aforementioned, only one study made use of VR glasses (Mohd et al. 2019). Further tools,
that were utilized, are controller, microphones, speakers, joysticks, an eye tracker, various
18
kinds of hand sensors, and the Kinect sensor (Microsoft Corporation, Redmond, WA, USA), a movement tracker that enables the user to control the software merely by the means of body- movement.
When it comes to the software integrated into the interventions, no identical software could be identified within the 11 records. However, the kinds of software that were made use of, consisted of game design software, telecommunication platforms and commercial video games and the so-called Reh@City software.
Inpatient. In terms of immersiveness, two out of six studies of the inpatient-setting reported the usage of immersive VR. Those were the studies by Lin et al. (2020) and Wiley et al. (2019). Since the study conducted by Wiley et al. (2019) was a literature review, they reported to have included interventions of all modes, meaning immersive, semi-immersive, and non-immersive VR. Due to the immersive nature, those two authors also mentioned VR glasses as in terms of hardware as part of their interventions. The other hardware devices used in the clinical setting, consisted primarily of game consoles such as the Xbox or Wii or PC- workstations with monitors. Additionally, tools such as the Kinect were utilized as well.
Concerning the software, two studies made use of the Joystim, a 3D manipulator, and others used software such as Reh@Task and Reh@City, BTs-Nirvana, and commercial video games.
Mixed. In the mixed setting, only non-immersive VRs were made use of. However, all six of the studies applied Tablets as their main hardware device. One study additionally reported the use of a computer and tools such as hand sensors were used. With regards to the software, the only ones mentioned were the Unity game engine (version 4.6) used for
software development and “RecoverNow”.
Table 5- Intervention Technology Characteristics
Criteria Home Inpatient Mixed
Immersive 1 2 -
Hardware
o VR-Device (glasses) 1 2 -
o Screen/monitor 4 1 -
o PC 6 1 1
o Tablet/ mobile device 3 - 6
o Games console (Wii, Xbox) - 2 -
19
Table 5 continuedCriteria Home Inpatient Mixed
Additional tools
o Controller/ Joystick 4 - -
o Kinect 1 2 -
o Microphone/speakers/webcam 5 - -
o Handsensor/Wristbands/smartgloves 4 - 3
Intervention design
Home. Six out of the 11 interventions in the home setting established cognitive rehabilitation as one of their overall intervention goals, followed by five interventions having upper extremity or motor improvement as their primary treatment target and one intervention explicitly mentioning mental well-being as a primary outcome. Additionally, the goal of two studies was not primarily the treatment but the recognition of multimodal affective states (Rivas et al., 2019) to create a system that automatically adjusts the exercises to a patients needs. Dodakian et al. (2017) presented a home-based telerehabilitation system, that can measure depressive symptoms.
The main content mode of the interventions consisted of games, which could take the form of serious games, exergames, videogames, etc. as indicated by all 11 of the home-setting studies. Other modes that were made use of, are an educating approach, exercises such as card sorting or the tower of Hanoi, and telecommunication with practitioners, nurses or therapists.
The duration of the interventions would range from 20-70 minutes, 5-7 times a week for 4-12 weeks. Further, different types of assistance were offered by the researcher team such as an initial introduction of the treatment modality and if necessary a caregiver could be included.
Only in the study of Calabro et al. (2018), a telecommunication system was integrated to implement videoconferences with the therapist to process and evaluate the progress. The interventions of Calabro et al. (2018), Dodakian et al. (2017), and Rai et al. (2017) included an online platform were both the therapist as well as the patient could monitor the process and the therapist could adjust the tasks.
Inpatient. In the inpatient-setting, three interventions focused on rehabilitation of
upper extremities and motor function, two on cognitive rehabilitation, and one on the patients
quality of life as their primary target. The modes of content provision were games, education,
and exercises. Telecommunication was not needed. The duration of a single session could
20
vary between 15-180 minutes and two to seven sessions were conducted within 1-8 weeks. In this category, it occurred that the VR intervention was implemented in combination with conventional therapy.
Mixed. Three interventions focused on upper extremity/ motor functioning, three emphasized cognitive rehabilitation, one quality of life, and another one operated as an assessment tool as the primary intervention outcome. The authors' Wall et al. (2018) demonstrate preliminary support for the feasibility of an App for cognitive assessment for patients with aphasia, which can be used for in- as well as outpatients.
Content-wise did three studies describe the application of a game mode, three presented exercises, and two mentioned telecommunication. With regards to the duration of the intervention, one session could last between 20-60 minutes, 3-7 times per week in 8-24 weeks. The longitudinal interventions of 24 weeks were part of Torrisi et al. (2019) and Maresca et al. (2019) studies. They introduced a two-phased intervention. The first half consisting of 12 weeks would be implemented as an inpatient/outpatient and the second half could be conducted from home.
In terms of assistance, three of the six mixed setting studies offered an initial introduction to the technology and training and one mentioned to include the caregiver, if necessary. Two studies integrated videoconferences twice a week with a therapist to monitor the process and if necessary adjust the tasks. And one study provided an online platform for the therapist and patient to monitor the progress and adjust the tasks.
Table 6- Intervention Design Characteristics
Criteria Home Inpatient Mixed
Primary Intervention Goal : Treatment
• upper extremity/ motor functioning 6 3 3
• cognition 6 2 3
• well-being 1 - -
• quality of life - 1 1
Assessment 1 1
Mode
• Game (serious game, exergame…) 11 3 3
• Education 2 1 -
21
Criteria Home Inpatient Mixed
• Exercise 3 3 3
• Telecommunication 3 - 2
Duration
• Minutes 20-70 15-180 20-60
• Sessions per week 5-7 2-7 3-7
• Weeks 4-12 1-8 8-24
Assistance
• Initial introduction 4 - 3
• Including caregiver, if necessary 2 - 1
• Video conferencing for process monitoring 1 - 2
• Online platform for process monitoring and task adjustment 3 - 1
Study outcomes
With regards to the second objective of this review, examining the effectiveness of VR interventions aimed at improving a) cognition and b) mental health among stroke survivors in the different settings, 17 out of 23 studies will be used for the quantitative synthesis.
Cognition
Both primary and secondary outcomes were included, measuring a participant's cognitive state. Those measurements would either assess the overall cognitive state or be specific towards e.g. attention, memory, executive function and spatial awareness
Home. When it comes to executive functioning, the authors Burdea et al. (2019) and
Fu et al. (2019) could identify significant gains due to their at-home intervention within their
data. The authors Chen et al. (2020) described that patients experienced enhanced cognitive
skills, as well as memory through playing the games of the intervention. Maier et al. (2020)
stated that their at-home Adaptive Conjunctive Cognitive Training (ACCT) might positively
influence attention and spatial awareness. Considering the technology that was used in those
studies, all of them were using a computer as well as a screen. Further, additional tools
operated with the hands were used such as controllers, joysticks, or gloves. All four studies
focused on games as their content mode. Other factors such as duration, software or assistance
differed.
22
Inpatient. The authors Kannan et al. (2019) noticed an improvement in balance as well as cognition due to the designed exergame intervention, which might reduce the motor- cognitive interference after stroke. The author Oh et al. (2019) described that a combination of VR with real instruments could be beneficial for cognitive functioning, as the client gets to practice daily activities in a simplified mode. In the meta-analysis by Wiley et al. (2019), outcomes revealed, however, that Virtual therapy would not be more effective than
conventional therapies as compared to the control groups when it comes to global cognition.
Kannan et al. (2019) and Oh et al. (2019) both described different technology systems, the one being a game console and the other being a workstation with a computer. The latter could also be found in the review conducted by Wiley et al. (2019). All three studies emphasized cognitive rehabilitation as one of their primary treatment goals. However, with regards to the other criteria of intervention characteristics, no overlap could be found
Mixed. The findings of the meta-analysis by Wiley et al. (2019) are in line with the findings of Lee et al. (2020) who suggested that the outcomes of their VR-intervention group did not differ from the control group. In Lee et al.s´ study, the primary outcome was to enhance motor function in outpatients that visited a local rehabilitation unit for their treatment. Contradicting to those outcomes are the findings of Torrisi et al. (2019) who demonstrated the effectiveness of their VRRS treatment for cognitive impairments post- stroke. Torrisi et al.´s intervention comprised, as earlier described, of a 2-phased intervention, including treatment as an inpatient followed by treatment at home, which specifically aimed at improving cognitive functioning. Torrisi et al.´s outcomes refer to the global cognitive level as well. Both, Lee et al. (2020) and Torrisi et al. (2019) made use of technological devices specifically designed for VR. Further, both researchers integrated game-like exercises in their intervention.
Apart from treatment interventions, did Wall et al. (2018) demonstrate preliminary support for the feasibility of an App for cognitive assessment for patients with aphasia, which can be used for in- as well as outpatients.
Mental health
In terms of mental health, the studies mainly included the symptoms of depression and anxiety. One study mentioned the survivor's mental well-being. All of the findings regarding mental health were secondary outcomes.
Home. The studies conducted by Burdea et al. (2019), as well as Maier et al. (2020),
provided support for the assumption that VR therapy at home might reduce depression in
23
stroke survivors. The authors Chen et al. (2020) mentioned in their qualitative study that patients “felt more socially connected after using the system”. All three interventions comprised monitors, tools for the hands such as controllers or wristbands and all ran in a game mode.
Regarding assessment tools, Dodakian et al. (2017) presented a home-based telerehabilitation system, that can measure depressive symptoms.
Inpatient. In the inpatient setting, Cano-Manas et al. (2020) indicated that a
combination of conventional treatment with a semi-immersive video-game approach might result in a positive effect when it comes to the “perception of pain/discomfort, sensation of anxiety/depression”. Those findings are in line with the research conducted by Lin et al.
(2020.) stating that supplementing VR training to early rehabilitation might lead to an improvement in a survivors mood state, which includes depression and anxiety. In both studies the Kinect sensor device was present. Additionally. Mental health was targeted in each study. Another factor prevalent in both studies was the implementation of conventional
therapy, to which the VR-therapy was added.
Mixed. The authors Maresca et al. (2019) concluded that the application of their intervention, might reduce depression and promote psychological well-being. They introduced a VR tablet system into their intervention, which started with inpatients, being discharged after 12 weeks but keeping the VR-tablet training for another 12 weeks.
Similarities and differences
The third research objective served to examine similarities and differences with regards to the characteristics and effectiveness of VR interventions between the different settings, which will be presented in the following.
Differences
Several differences between the three location settings could be derived from the records (see Table 7). Concerning the technology, one of the main differences can be seen in the hardware that is used. In the home settings computer-based workstations, in the inpatient setting game consoles, and the mixed setting tablets were most prevalent. Further, additional tools were mentioned more frequently in the home-setting than the other two.
In terms of the overall intervention goal, two Assessment tools, one for depression
(Dodakian et al., 2017) and one for affective states (Rivas et al., 2019) were introduced and in
24
the mixed setting one for cognitive assessment (Wall et al. 2018). For the inpatient setting, no such thing is available as indicated by the records.
In the home setting, all four content modes could be identified, games, education, exercise as well as telecommunication. In the inpatient setting, telecommunication was missing, which however might be expected. In the mixed setting, no education mode was found in the records. The minimum and maximum duration of the period, the intervention was offered, differed between the settings as well, with an inpatient setting starting by only one week, followed by the home setting with four weeks and the mixed setting with 8 weeks. The maximum duration differed as well, as a maximum of 8 weeks were given in the inpatient setting, followed by the home setting with 12 weeks and a total of 24 weeks by the mixed setting.
As far as conclusions can be drawn with regards to the effectiveness, in the home setting, studies preliminarily indicated that VR might be beneficial for cognitive
rehabilitation. In the inpatient and mixed setting, however, results deviated. Results of the meta-analysis conducted by Wiley et al. (2019) and an RCT by Lee et al. (2020) concluded that VR is not necessarily more effective than conventional therapy, as measured by means of a control group. However, a more recent RCT in the inpatient setting pleads for the more convenient usability of VR for cognitive rehabilitation.
Only in the inpatient setting, assumptions were drawn that VR might be related to an improvement of mental health concerning anxiety. In general, mental health was addressed in all inpatient studies, but only partly in the home-setting. Further, VR was seen as a
supplementing factor in the inpatient settings, whereas in the home setting it partly would stand by itself.
Table 7- Differences between the settings
criterium home inpatient mixed
hardware PC Game console Tablet
tools Used very
frequently
Barely used, but if, Kinect Barely used, but if, Handtools Overall
Intervention goal
assessment No assessment assessment
Content mode All 4 modes No telecommunication No education
25
criterium home inpatient mixed
duration 4-12 1-8 weeks 8-24 weeks
Effectiveness
cognition Preliminary effective
Meta-analysis says not more efficient than the control group, but newer RCT pleads for possible positive impact
Not more
efficient than control group, Mental well-
being
Positively influence depress
Pos. influences depress/anxiety
-VR in early rehabilitation might affect mood including Anx. & depression
Pos. influences Depress.
Not each study Mental health targeted in each study - Only VR or
additionally
VR as a supplementing factor
Similarities
A summary of the similarities can be found in Table 8. Similarities about the hardware were computer-based workstations that could be found in each setting, even when they were not the main choice of hardware. Further, in all settings, the overall intervention goal of treating upper extremities/ motor function or cognitive functions could be found. All settings included games or exercises as content modes.
An overlap could be seen in the duration per session, as in each category sessions of 20-60 minutes could be found, and the number of sessions per week could also go up to seven days per week in each setting. Especially with regards to the home and mixed setting, an overlap could be seen in the assistance provided. Both settings checked all four common criteria of an initial introduction, inclusion of the caregiver, videoconferencing and an online platform for process monitoring and task adjustment available for both, therapist and client.
Table 8 - Similarities between the settings